Synergistic Effects of BubR1 and p53 Deficiency in Tumor Formation

Transcription

1 University of Tennessee, Knoxville Trace: Tennessee Research and Creative Exchange Masters Theses Graduate School Synergistic Effects of BubR1 and p53 Deficiency in Tumor Formation Walter Guy Wiles University of Tennessee - Knoxville Recommended Citation Wiles, Walter Guy, "Synergistic Effects of BubR1 and p53 Deficiency in Tumor Formation. " Master's Thesis, University of Tennessee, This Thesis is brought to you for free and open access by the Graduate School at Trace: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Masters Theses by an authorized administrator of Trace: Tennessee Research and Creative Exchange. For more information, please contact trace@utk.edu.

2 To the Graduate Council: I am submitting herewith a thesis written by Walter Guy Wiles entitled "Synergistic Effects of BubR1 and p53 Deficiency in Tumor Formation." I have examined the final electronic copy of this thesis for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Master of Science, with a major in Biochemistry and Cellular and Molecular Biology. We have read this thesis and recommend its acceptance: Ranjan Ganguly, Ana Kitazono (Original signatures are on file with official student records.) Sundar Venkatachalam, Major Professor Accepted for the Council: Dixie L. Thompson Vice Provost and Dean of the Graduate School

3 To the Graduate Council: I am submitting herewith a thesis written by Walter Guy Wiles IV entitled Synergistic Effects of BubR1 and p53 Deficiency in Tumor Formation. I have examined the final electronic copy of this thesis for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Master of Science, with a major in Biochemistry and Cellular and Molecular Biology. Sundar Venkatachalam Major Professor We have read this thesis and recommend its acceptance: Ranjan Ganguly Ana Kitazono Accepted for the Council: Anne Mayhew Vice Chancellor and Dean of Graduate Studies (Original signatures are on file with official student records)

4 Synergistic Effects of BubR1 and p53 Deficiency in Tumor Formation A Thesis Presented For the Masters of Science Degree The University of Tennessee, Knoxville Walter Guy Wiles IV May, 2006

5 Dedication: This thesis is dedicated to my wife, April Wiles, and my parents Guy and Paula Wiles and Ron and Gail Williams for their love, their belief in me, and encouraging me to succeed. ii

6 Acknowledgements: I wish to thank all those who have offered their knowledge, time, and support to this thesis. I would like to thank Dr. Venkatachalam for his guidance and friendship during this work. Finally, I would like to thank my family and friends without whose support would have made this endeavor much more difficult. iii

7 Abstract: Regulation of cell proliferation within an organism is a necessary and complex process involving several proteins serving as controls at various cell cycle phases known as cell cycle checkpoints. One of these checkpoints, the mitotic spindle checkpoint, controls the advancement from metaphase to anaphase during mitosis and monitors the proper attachment of the microtubules to the kinetochore. The spindle checkpoint protein, BubR1 is a protein kinase that localizes to the kinetochores to monitor proper microtubule attachment. BubR1 is able to inhibit the anaphase promoting complex (APC) and delay the onset of anaphase. Concurrent with its role in regulating cell cycle, mutations in BubR1 have been observed in various human cancers. In this work, we examine the effects of BubR1 deficiency at the cellular and organism level using an inbred mouse strain that is deficient for the BubR1 gene expression. We showed that the complete loss of BubR1 resulted in an embryonic lethal phenotype and this phenotype could not be rescued in a p53 deficient background. Additionally, mice heterozygous for BubR1 showed a decreased life expectancy and an increased incidence of tumorigenesis. Furthermore, our data indicates that the loss of BubR1 synergizes with p53 deficiency to increase the susceptibility of cancer formation in mice and alters the tumor spectrum of the p53 deficient mice. Finally we show that the BubR1 deficiency increases cellular growth kinetics and transformation potential of MEFs derived from compound BubR1; p53 mutants. This data provides insight into the importance of BubR1 in the prevention of tumorigensis and its role as a checkpoint protein. iv

8 Table of Contents: I. Introduction 1 A. Cancer Overview and Statistics 1 B. Cancer and the Cell Cycle 2 C. Cell Cycle Checkpoints and Mitosis 4 D. Role of p53 as a Checkpoint Protein 8 i. Overview 8 ii. Activation of p53 10 iii. Physiological role of p53 11 iv. Regulation of p53 activity 13 v. Effects of p53 inactivation 15 F. Role of BubR1 as a Checkpoint Protein 16 i. Overview 16 ii. Functional role of BubR1 19 iii. BubR1 activation and the mitotic spindle checkpoint 21 iv. Additional functions of BubR1 22 v. Effects of BubR1 inactivation 23 G. Rational for Generating BubR1/p53 Compound Mutants. 25 II. Materials and Methods 27 A. Generation of BubR1 Mutant Mice in a p53 Deficient Background 27 B. DNA Extraction 28 C. BubR1 Genotyping 28 D. Generation of Mouse Embryonic Fibroblasts (MEF) 29 E. Cell Proliferation Assays 30 F. Colony Forming Unit Assay (CFU) 30 G. Cell Cycle Analysis 31 H. Animal Husbandry 31 I. Southern Blotting 32 III. Results 35 A. Characterization of the BubR1 Gene Trap Insertion Site 35 B. BubR1 Deficiency Effects Embryogenesis 35 C. Generation of BubR1/p53 Compound Mutants. 38 D. BubR1 Deficiency Enhances Tumor Formation in p53 Mutant Mice 41 E. Gene Dosage Effect of BubR1 During Tumorigenesis 48 F. BubR1 Deficiency Enhances Cell Proliferation and Cell 48 Transformation Potential IV Discussion 52 References 56 v

9 List of Tables: Table Page 1. Embryonic lethality of BubR1 nullizygotes p53 deficiency does not rescue the lethality of the BubR1 nullizygotes p53 deficiency is unable to delay the lethal phenotype observed in BubR1 42 nullizygotes. vi

10 List of Figures Figure Page 1. The stages of mitosis The mitotic spindle checkpoint The BubR1 protein Southern blot results indicate the p53 genotype The BubR1 gene and the location of the PCR primers PCR indicates the BubR1 genotype Survival of the wild type, heterozygous, and compound heterozygous 43 mutant mice. 8. Survival of BubR1 mutants in the p53 nullizygous background Tumor spectrum of the single and compound heterozygous mice Tumor spectrum of the single and compound nullizygous mice BubR1 and p53 deficiency has a synergistic effect on cellular proliferation BubR1 and p53 deficiency has a synergistic effect on the transformation 51 potential of MEFs vii

11 Nomenclature cpm/ml counts per minute per milliliter µl microliter µm micromolar mg milligrams mm millimolar ng nanograms Abbreviations APC/C anaphase promoting complex/cyclosome CDC cell division cycle/control CDK cyclin-dependent kinase CFU colony forming unit CIN chromosomal instability DMEM Dulboc s modified eagle media DNA deoxyribonucleic acid E10.5 embryonic day 10.5 E13.5 embryonic day 13.5 FBS fetal bovine serum G1 gap phase 1 G2 gap phase 2 M mitosis MCC mitotic checkpoint complex MEF mouse embryonic fibroblast NLS nuclear localization signal PBS phosphate buffered saline Pen/Strep penicillin/ streptomycin RNA ribonucleic acid ROS reactive oxygen species S synthesis TBP TATA binding protein TAF TBP associated factors viii

12 I. Introduction: Cancer Overview and Statistics: Cancer is a disease of uncontrolled cell growth and division (1). In this disease, a single cell, through mutation, becomes unable to monitor or control its proliferation. This unchecked proliferation grants a growth advantage to the cell, and with time, allows further mutations to accumulate within the cells created by the original cell, called clones. These accumulated mutations will alter the clones in such a way that they are able to escape regular growth control signals and, in some instances, develop the ability to escape their primary site, referred to as a tumor, and spread throughout the body of the organism (1). Mortality is often due to the fact that the rapidly dividing cells fail to differentiate and perform the function they are designed to. Alternatively, these masses can grow, affecting the normal functioning of adjacent organs or they can usurp the blood supply to the normal tissue causing the organs near the mass to fail (2). Mammals have evolved a variety of mechanisms to control/regulate cellular proliferation. However, over time, these control mechanisms can become compromised due to accumulation of genetic changes that result in uncontrolled cell growth. Cancer is a disease prevalent world wide, and the loss of life and costs of treatment have remained consistently high over the last 50 years, even though medical advances have lowered the mortality and cost of many other commonly seen health threats. Cancer can effect any organ or tissue type, but for the purposes of this study, we will examine the trends of the disease overall and the statistics of colon/rectal cancer. According to the American Cancer Society, in 2005, cancer was the second highest cause 1

13 of death with 22.8% of all deaths being attributed to cancer or its complications. Cancer deaths are second only to heart disease which is the leading cause of death with 28.5% of all deaths. The difference between the two, 5.7%, is relatively small especially when one looks at the separation between the second and third leading cause of death, cerebrovascular diseases, which accounts for only 6.7% of all deaths. This information alone would be enough to promote a desire to understand this disease, but with the actual number of deaths indicated, the need becomes more pressing. It is estimated that in the US alone, 295,280 men and 275,000 women died from cancer in The leading type of cancer for these people is lung cancer (31%), though cancers of the colon/rectum will cost 10% of both men and women their lives, and are the third leading cause of deaths from cancer. Also, it is estimated that 710, 040 men and 662, 870 women in the US were diagnosed with new cases of cancer in Again, cancers of the colon/rectum will be the third most common cancer to be diagnosed. According to these statistics, one out of two men and one out of three women will develop cancer over the course of their lifetime with one out of 17 of those, for both genders, being cancers of the colon. Also it has been estimated that the cost of cancer in 2003 was $189.5 billion. Thus, it becomes apparent that cancer is a disease in which discovering effective treatments to prevent or reduce disease occurrence is an important step in promoting people s health (2). Cancer and the Cell Cycle: Since cancer is a disease caused by uncontrolled cellular division, it is first necessary to examine the way in which cell division occurs under normal conditions. Cellular division consists of four stages. These stages are G1 (gap phase 1), S 2

14 (synthesis), G2 (gap phase 2), and M (mitosis). In G1, the cell has just finished a round of cellular division and requires time to grow and obtain the nutrients and build the proteins necessary to begin the next round of cellular division. Once the required needs are met, the cell will enter S phase where the cell will replicate its DNA. DNA synthesis is followed by a second gap phase where, again, the cell makes sure that it has time to gather the necessary machinery to allow accurate and timely division as well as to assure the integrity of the cell s genetic material. Finally, during the M-phase, the cell undergoes mitosis and completes cellular division. The cell cycle is conserved in all dividing cells and is regulated by proteins known as cyclins and their partners, the cyclindependent kinases. (CDK). The CDKs are found constitutively in the cell during all of the cell cycle stages. Cyclins on the other hand, are only expressed during the stage in which they associate with their CDK partners. The cyclins then are ubiquitinated and destroyed by the proteosome to allow progression into the next stage of the cell cycle (1, 3, 4). To facilitate the ubiquitination and subsequent destruction of these products, they contain a conserved amino acid sequence of RxxLxxxxS. This sequence, known as the destruction box sequence, is found in many cyclins as well as several other cell cycle proteins that are regulated in a cell cycle dependent manner, such as BubR1. One way in which normal cells develop a cancerous phenotype is through the improper activation or production of these cyclins controlling the passage of the cell from one stage of the cell cycle to the next (5). 3

15 Cell Cycle Checkpoints and Mitosis: To assure the proper progression through the cell cycle, and to prevent the progression to the next stage before the requirements of the previous stage are met, the cell cycle has safeguards known as checkpoints. These checkpoints are found at critical junctions of the cell cycle and allow the cell cycle to be arrested if there is DNA damage or if the cell cycle progression stalls due to aberrant conditions. Upon activation of these checkpoints, the cell will cease the cycle at whichever point it is in until the event that activated the checkpoint is corrected for, allowing the cell cycle to continue. If the situation is too dire, extended activation of the checkpoint will promote apoptosis. The three commonly described cell cycle checkpoints are the checkpoints found at the G1/S transition, which allows the cell to begin DNA replication, the G2/M transition, which assures that the DNA was replicated without error, and the mitotic spindle checkpoint which helps assure that there is accurate chromosome segregation (4, 6). A fourth cell cycle check point is the intra-s checkpoint, which is activated when the replication fork encounters DNA damage during replication. This checkpoint enhances genomic stability by slowing DNA synthesis to allow time for DNA repair. (7) The G1/S checkpoint is activated by DNA damage and will stall the progression into S-phase by preventing the firing of the replication origins or progression of the replication fork until the damage is repaired. This G1/S arrest allows for repair of the template DNA and prevents DNA damage from becoming a fixed mutation in the cell for subsequent divisions. The major players of the G1/S checkpoint include the proteins ATM, ATR, p53 and p21 (8). In this checkpoint, DNA damage is sensed by ATM Kinase and p53 protein levels are increased. The p53 tumor suppressor gene protein, 4

16 then activates several gene products leading to delay of the cell cycle progression (9). Once the DNA damage is repaired, the checkpoint signal is reduced through the destabilization and degradation of p53, and the cell is able to continue on to and through S-phase. After DNA replication, there is a second DNA damage checkpoint to assure that both the template and the replicated DNA are undamaged. This occurs after replication but before progression into mitosis and serves to assure that the daughter cells produced both have DNA that is identical to the parent cell (4). The major components of the G2/M checkpoint include Chk1, Chk2, and p53 proteins. If DNA damage is detected during the G2-M transition phase, ATM or ATR will activate Chk1 and Chk2. These, in turn, will phosphorylate and inactivate the phosphatase Cdc25c (10, 11). The role of Cdc25C is to remove an inhibiting phosphorylation on Cdc2 (tyrosine-15). This inhibition of Cdc2 stops the cell cycle from progressing into M-phase until the DNA damage is repaired. The p53 protein is also activated by ATM or ATR in response to DNA damage and serves a role similar here as in the G1/S checkpoint by inducing the expression of p21 that arrests the cells at the G2/M phase. It is only after the progression through these two checkpoints that the cell enters M-phase and subsequent cellular division (6, 12). Mitosis is a multi-step process in which the parent cell must condense its DNA into paired chromosomes, align the chromosomes along the metaphase plate, accurately separate the sister chromatids, and finally reform the nuclear membrane and de-condense the DNA. The first step or stage of mitosis is known as prophase, and it is here that the nuclear membrane breaks down and the DNA begins to condense. At this time, the sister 5

17 chromatids are secured together by proteins called cohesins, serving as molecular glue so that they can eventually be divided equally between the two daughter cells. There is then an intermediate stage known as prometaphase, where the chromosomes have condensed and are free in the cytoplasm and the centrioles have reached the poles and begin building microtubules. These microtubules will extend out into the cytoplasm and, upon making contact, will bind to the kinetochore of the chromosomes. Eventually, at metaphase, the microtubules will align the chromosomes up in the middle of the cell with the microtubules attached to either side of the kinetochores. This places tension across each kinetochore, signaling bipolar attachment. Once aligned across the metaphase plate, the chromosomes then go through anaphase where the sister chromatids separate and move to the poles. Once congregated at the poles, the chromosomes de-condense and the nuclear membrane reforms around each set during telophase (figure 1). The cellular contents then divide in a process known as cytokinesis and the two daughter cells are now formed (1, 4). It is during mitosis that we find the third cell cycle checkpoint, known as the mitotic spindle checkpoint. In order to assure proper separation of the sister chromatids to the daughter cells, the chromosomes must be aligned on the metaphase plate and each of the chromatids must be attached to microtubules originating from opposing spindle poles. Since the microtubules employ a search and capture method of attachment, progression must be delayed until all of the chromosomes are attached and tension is generated across the kinetochore, signaling bipolar attachment (13, 14). It has been shown that the mitotic 6

18 Prophase Prometaphase Metaphase Anaphase Telophase Cytokinesis Figure 1: The stages of mitosis. This diagram shows the stages of mitosis. In prophase and prometaphase the nuclear membrane breaks down and the DNA condenses. It is here that many of the mitotic spindle checkpoint proteins will bind to the kinetochores of the chromosomes. Once condensed, microtubules will attach to the chromosomes and line them on the metaphase plate (metaphase) and upon alignment the sister chromatids will be separated and pulled towards the poles (anaphase). Finally the chromosomes will reach the poles and de-condense and the nuclear membrane regenerated and the cell will divide (telophase and cytokinesis). (images from 7

19 spindle checkpoint signal is activated even if only a single kinetochore remains unattached (15). The major components in the mammalian mitotic spindle checkpoint include Mad1 and Mad2, Bub1, BubR1, and Bub3 (16). These proteins all localize to the kinetochores of unattached microtubules during prometaphase and become activated during metaphase, where they create a complex known as the Mitotic checkpoint complex (MCC), along with the binding of Cdc20. Once the MCC is formed and activated, believed to be through various phosphorylation events by the other checkpoint proteins, the MCC then prevents the activation of the anaphase promoting complex or cyclosome (APC/C) (17). The APC complex directs the ubiquitination of several components including the protein securin. This degradation of securin allows its bound partner, separase, to be released and cleave Scc1, a protein involved in keeping the sister chromatids together during the early stages of mitosis. This frees the sister chromatids from each other and allows the separation and movement of the chromatids to the spindle poles. Also, the APC will ubiquitinate several of the mitotic proteins marking them for degradation, allowing mitosis to end and the cell cycle to start anew (18, 19) (figure 2). Role of p53 as a Checkpoint Protein: Overview: As mentioned above, cancer is a disease stemming from uncontrolled cellular proliferation. In order to lose control of proliferation mutation in key cell cycle components or their checkpoints are necessary. It is interesting to note that the most commonly mutated gene in human cancers is p53. As previously stated, p53 serves as a checkpoint in both the G1/S and the G2/M phases. This makes p53 an ideal protein for 8

20 Figure 2: The mitotic spindle checkpoint. This diagram shows the role of BubR1 as well as other spindle checkpoint proteins in the activation of the mitotic spindle checkpoint. In prometaphase, unattached kinetochores will allow the localization of multiple checkpoint proteins including BubR1 and Mad2. Once localized and activated (believe to be through phosphorylation events) these proteins interact with Cdc20 to inhibit the APC. Once attachment and tension is signaled these checkpoint proteins become inactive and the APC is no longer inhibited and anaphase begins (20). 9

21 cancer study since insights on the mechanisms of p53 mediated cell cycle control could serve as the foundation for new or more effective cancer treatment strategies. The gene product of p53 is a protein of ~53 kda that is made up of three domains. These domains include an activation domain at the amino-terminal spanning residues 1-42, a central core containing the sequence-specific DNA-binding domain spanning residues , and finally a carboxyl-terminal, multi-functional domain spanning residues Each of these domains connects through a flexible linker sequence (21). In the protein, there are five evolutionarily conserved regions with the first one being in the N-terminal activation domain and the other four found in the central core. Multiple phosphorylation sites are found in both the N-terminal and C-terminal domains, which are used to regulate the activity of p53. (22) Activation of p53: After exposure to DNA damaging agents, multiple forms of DNA damage are induced that include double strand breaks and base alterations. This damage leads to the activation of p53 through the actions of multiple kinases that phosphorylate p53 at multiple sites. The kinases involved in this response include JNK, ATM and ATR (10, 11). In addition to the phosphorylating events at serines 15, 20, and 37, there is also a dephosphorylating event at serine 376 which opens a new site in p53 for interaction with the scaffold protein (8, 23, 24). Phosphorylation of the above serines, stabilize the protein and inhibit its interaction with the negative regulator protein Mdm-2. Once stable, p53 is transported into the nucleus via phosphorylation of serine 392 by CK II where it tetramerizes and exhibits its transcriptional activity (25-27). Unlike the rapid 10

22 response seen to ionizing radiation, DNA damaged induced by UV has a slower response time. Since the damage caused by UV is most often the formation of bulky lesions such as thymidine dimers there is no immediate detection of the lesion until the replicative or transcriptional machinery stalls upon reaching the lesion. Once the damage is detected; the stalling of RNA polymerase II causes serine 33 on p53 is phosphorylated by the CDK Activating Kinase (CAK) from TFIIH. Additionally, ATR activates Chk I and Chk II which phosphorylate serine 20 while ATR directly phosphorylates serine 15 and 37 (28). As seen above in response to ionizing radiation, these phosphorylation events, lead to the stabilization of p53. Once stabilized, p53 is again phosphorylated on serine 392, and transported into the nucleus where it tetramerizes and induces the transcription of the DNA damage response genes (29). The various phosphorylations of p53 lead to the alteration of the protein structure and expose the central core s sequence-specific DNAbinding domain as well as allowing the N-terminal activation domains to interact with other transcriptional factors such as TATA box-binding protein (TBP) and its associated factors (TAFs) (30). Finally, since p53 accumulation will sequester the cofactors involved in transcription, those genes that do not contain a p53 specific binding site will not be transcriptionally active until the level of p53 is restored to its base (31-34). Physiological role of p53: The activation of p53 leads to three outcomes for the cell; cell cycle arrest, cellular senescence, or apoptosis. By keeping with these outcomes, p53 is able to protect the genome from accumulating mutations. However, since these effects prevent cellular growth, p53 must be tightly regulated to prevent wasteful delay, arrest and/or 11

23 unnecessary apoptosis. In many organisms, the extensive loss of cells due to up regulation of p53 can lead to problems which can put the entire organism at risk. On the other hand, if p53 is too tightly regulated due to an over-expression of its negative regulators or an under-expression of the gene itself, the stability of the genome can become uncertain due to the lack of p53-dependent checkpoint activation. Once activated, the results of the activation must be decided as to whether the cell undergoes cell cycle arrest or apoptosis. This decision is based on several factors including the cell type, severity of damage, and the time that p53 activity was induced. Most of the genes induced by p53 lead to several outcomes that include cell cycle arrest, apoptosis, and cellular senescence. An example of p53 induced cell cycle arrest is the induction of the p21 gene whose product promotes cell cycle arrest in G1/S through binding and inhibition of the CDKs necessary for cell cycle progression. Activated p53 has been shown to bind to the p21 promoter. This binding increases transcription of the gene product and p21 will then form part of a quaternary complex with PCNA and the cyclin/cdk complex. This quaternary complex will inhibit the cyclin/cdk activity, arresting the cell cycle in G1. In addition to p21, after ionizing radiation, p53 will also bind to the promoter of GADD45 (growth arrest and DNA damage inducible gene) and activate it. This protein will then interact with PCNA, a replication and repair factor and inhibit it. With the inhibition of PCNA the cell will be unable to enter S-phase and the cell cycle comes to a halt. Also, it has been shown that GADD45 is able to alter the chromatin accessibility which could inhibit progression into S-phase by maintaining the DNA in an inaccessible form to other replication factors until the damage is repaired (22, 26). In some cell types, activation of p21 by p53 leads to prolonged cellular arrest. 12

24 Dependent on the damage or the cell type, the activation of p53 may induce apoptosis (35). This pathway begins similar to the cell cycle arrest pathway in that p53 is activated, stabilized, and transported into the cell nucleus. However, from here things become a bit less characterized. While it is known that p53 mediated cell cycle arrest requires the protein s transcriptional activity, initial reports seemed to indicate that the transcriptional activity may not be necessary and it was suggested that p53 may have both a transcription dependent and independent pathway to induce apoptosis. (22) However, a later work showed that the induction of apoptosis through p53 is a three step process which requires p53 to be transcriptionally active. From this study it was determined that upon the activation of p53, a subset of genes known as the p53 inducible genes (Pigs) are promoted. The transcription of these genes, which code for several proteins involved in regulation of reactive oxygen species (ROS), will lead to an increase in the level of ROS. This increase of the ROS level will then work to decrease the stability of the mitochondrial membrane that leads to Cytochrome-C release. Once released, Cytochrome-C will interact with several other apoptotic proteins ultimately leading to the activation of the caspase cascade that will induce genomic degradation and membrane blebbing, hallmarks of apoptosis. Regulation of p53 activity As seen from the information presented already, the regulation of p53 is tightly controlled and the activating cascade requires several players to properly stabilize, activate and localize the protein so that it may exert its effect. In addition to these activating proteins, p53 is also regulated on its transcriptional and translational levels. 13

25 Transcriptionally, p53 is induced in response to the stress factors such as AP-1, NF-kB and Myc/Max as well as the products of YY1 and NF1. These transcription factors will bind to the promoter of p53 and promote transcription of the gene (36). The p53 gene is transcriptionally repressed by the Pax transcription factor family (37). Also, the viral protein Tax and the over expression of c-jun have been shown to repress transcription of p53(38, 39). During the translation of the p53 mrna, p53 has been shown to be able to repress its own translation through the 3 untranslated region (UTR) in human and the 5 UTR in murine p53. These UTR sequences can form stable secondary structures which can repress translation through interaction with the RNA binding factors. Though p53 is regulated both in its transcription and translation, the majority of its regulation is seen post-translationally. In the cell, cytoplasmic p53 has a short lifespan of only a couple minutes. The protein p53 can induce the transcription of proteins which attenuate its activity serving as an auto-regulatory feedback. An example of this is the mdm-2 gene, whose protein product binds to p53, preventing its nuclear localization and promoting its degradation, is a downstream target of p53 (22, 25, 31). Thus p53 upregulates its own inhibitory factors, thereby assuring that p53 activity will diminish quickly after the DNA damage is repaired (40). Cytoplasmic p53 is bound to the protein MDM-2 and ubiquitinated, targeting the protein for degradation by the 26s proteosome (41, 42). In addition to this degradative pathway to regulate p53, the localization of p53 is another way of controlling p53 activity. For p53 to function, it must first be able to enter the nucleus of the cell. To do this, p53 has a nuclear localization signal which must be accessible to recognition sites on the nuclear membrane and so one method of regulation is via the binding of other proteins which hide the signal. One example of this 14

26 is the binding of p53 to Mdm-2 to prevent access of the localization domain. Also, once imported into the nucleus, p53 may bind to MDM-2 and be subsequently exported. By controlling the degradation or localization of p53, the cell assures that any protein present in the cell remains at sub-active levels until activation and stabilization of the protein is needed (22, 25). Effects of p53 Inactivation: With p53 serving such an important role in maintaining genomic stability and providing the means to arrest cell cycle progression or promote apoptosis, it is no small wonder that mutation of p53 or the complete loss of the gene is commonly seen in cancer. The most commonly seen mutations of p53 are found in the DNA binding regions of the protein. These mutations all serve to prevent p53 from inducing the transcription of cell cycle arrest or apoptotic genes (31). Additionally, mutations of the NLS region of the protein have been characterized. It has been observed that mutation of the NLS region or alteration of the residues lysine 305 and arginine 306 leads to the cytoplasmic localization of p53 and its inactivation (25). Finally, mutation of either p53 or MDM-2 to prevent their dissociation will lead to a substantial reduction of inducible p53 which will prevent p53 mediated cell cycle arrest or apoptosis (34). In all of these mutations, the inactivation of p53 leads to increased proliferation, genomic instability, and the accumulation of mutation due to the loss of cell cycle arrest and apoptosis. 15

27 Role of BubR1 as a Checkpoint Protein: Overview The protein, BubR1, is a part of the mitotic spindle checkpoint which serves to assure anaphase delay until the chromosomes are properly aligned on the metaphase plate with proper microtubule attachment and tension along all the kinotechores. The human BubR1 gene is located on chromosome 15q15 and its product is considered to be one of the major players in the prevention of chromosomal instability (CIN). While BubR1 mutations are not prevalent in all cancers, studies have shown that mutations of BubR1 do occur in some colorectal cancers. (43) In addition, chromosomal rearrangements of 15q15 have been observed in some leukemias. These rearrangements may promote the inactivation of BubR1 as a checkpoint control protein and increase the likelihood of tumor progression (44). BubR1 is a gene found in mammals but not in yeast and has homology to both yeast genes BUB1 and MAD3, hence the names BubR1 (bub1-related). BubR1 was first identified in a study by Davenport et al. using differential expression of normal and leukemic mouse thymocytes. In this study, a homologue to the yeast checkpoint protein Bub1 was discovered and this novel family member, named mbub1b, was shown to have 40% sequence similarity to murine Bub1a over 4 extended domains. To better understand this novel protein, Davenport et al. used the clone fragment B13 from their differential expression studies to isolate a 3,647 base pair cdna from the 16 day mouse embryo cdna library and determined some of the properties of this new protein (44). As mentioned above, the BubR1 gene transcribes a message of 3,647 base pairs making up 23 exons with a promoter located 1,368 base pairs upstream of exon 1. It was also found that a GC-rich region rests 150 base pairs upstream of exon 1. This 16

28 translates to a protein with 1,052 residues across four major domains. These domains are a Ken-box motif, a signaling sequence for degradation, (residues 26-28), a Mad3-like region (residues ), a Bub3 binding domain (residues ) and a kinase domain (residues ) (45). The kinase domain of BubR1 is atypical in that the kinase domains of Bub1, BubR1 and the yeast Bub1 are more closely related to each other than to other kinase domains. Even with this being the case, BubR1 contains significant alterations which mask its relationship to other kinases outside of the bub family. An example of this is the substitutions of Bub1 s Glutamate 127 and Aspartate 170 with BubR1 s Asparagine and Arginine (44). These alterations are explained in Davenport s work as proofs that BubR1 has different substrates than Bub1. Along with these domains, BubR1 has a destruction box sequence immediately following its aminoterminal domain, (residues in the mouse Bub1b) and two sequences that could serve as nuclear localization signals (residues and ) (44). Also, later work showed that BubR1 contained a dileucine motif (residues ) this motif is commonly seen as an adaptor beta chain recognition motif (46). The destruction box sequence, RSSLAELKS, is located between the Mad3-like and the Bub3 binding domains and serves to promote degradation of the protein through ubiquitination of the lysine residues targeting the protein for destruction by the proteosome. This destruction box allows the degradation of BubR1 in a cell cycle-dependent manner (44, 47). Along with these various sequences and within these domains are several phosphorylation sites. These sites are hypothesized to serve as activators and regulators for BubR1 activity in the event of mitotic spindle checkpoint activations (figure 3). 17

29 N-Terminal Domain Kinetochore Binding Domain Bub3 Binding Domain Kinase domain DB NLS Figure 3: The BubR1 protein. This diagram shows the BubR1 protein and its evolutionarily conserved domains as well as their role in the activity of BubR1. The Destruction box (DB) is located just past the N-terminal domain of the protein (containing its Mad3 like region and Ken box motif). The nuclear localization sequence (NLS) is located within the kinetochore binding domain. 18

30 Functional role of BubR1: As implied by its multiple structural domains and motifs, the role of BubR1 in the cell is complex. The exact function and binding partners as well as the kinase substrates of this protein have been sought out since it was first isolated in 1998 (43, 44). Many researchers have spent years on this gene and while strides have been made and some of its key roles determined, there is still a great deal to learn about this gene and its product. The isolation and study of BubR1 done by Davenport determined that the product was expressed in all dividing cells and highly expressed in the spleen and thymus. This work showed that expression of this gene peaked around the G2/M phase of the cell cycle and due to its homology to Bub1, it was determined that BubR1 has a distinct role in the mitotic checkpoint. This finding was confirmed in a separate study by Cahill et al. who examined 19 colorectal cancer cell lines that exhibited chromosomal instability (CIN) and found that in two of them there was a mutation of the human bub1 homologue BUBR1(43, 44). One of the first studies to determine BubR1 s binding partners and regulation was done by Chan et al examining the interaction of BubR1 with the centromeric motor protein E (CENP-E). It was found that while BubR1 levels were lowest in G1, it steadily increased towards mitosis. During Mitosis, BubR1 was hyperphosphorylated and it was found that this hyper-phosphorylation increased the kinase activity of the protein (48, 49). Their studies also showed that BubR1 was capable of binding to CENP-E in the cytoplasm of interphase cells through interactions of 641 residues at its C-terminal. Furthermore it was observed that BubR1 could localize to the kinetochores of unaligned chromosomes during prometaphase. This localization would 19

31 occur after the localization of CENP-F but before the localization of CENP-E suggesting that BubR1 may phosphorylate substrates on the kinetochore to allow the binding of CENP-E. Because it was accepted that BubR1 had a distinct role in the mitotic checkpoint and this discovery that BubR1 interacted with CENP-E, a motor protein, it was suggested that the BubR1/CENP-E complex served as a mechanosensor for the unaligned chromosome (49). Later experiments were able to show that BubR1 was an essential component of the mitotic checkpoint though CENP-E was not. Additionally, the kinase activity of BubR1 was examined and it was observed that while BubR1 had no detectable kinase activity in either interphase cells, or in mitotic checkpoint activated cells, through addition of the drug nocodazole, BubR1 showed a high level of kinase ability both for itself as well as a number of exogenous substrates using in vitro kinase assays. It was determined that one role of BubR1 was to prevent normal mitotic exit until the chromosomes had aligned properly and this arrest was through specific interactions with the Anaphase Promoting Complex (APC). In addition to this role in the mitotic checkpoint activation, it was confirmed that BubR1 served to monitor functions specified by CENP-E as a tension checkpoint (48). In 2001, in a study by Skoufias et al. it was confirmed that one of the major roles of BubR1 was to sense tension across the chromosome s kinetochore. It was shown that in the event of tension loss while microtubule attachment was maintained, Bub1 and BubR1 returned to the kinotechore to delay anaphase until tension was restored. It was suggested that BubR1 and another mitotic checkpoint protein Mad2 were parts of two distinct checkpoints, one sensing tension and the other sensing microtubule attachment (50). However, this model was later proven to be incorrect through a study by Shannon et al. when it was shown that in 20

32 cell lines under hypothermic conditions, the major cause of prometaphase delay due to lack of tension, that the mitotic delay could be attenuated either through the addition of antibodies against Mad2 or BubR1. Additionally, it was observed that the addition of both antibodies did not accelerate the progression into anaphase. This showed that BubR1 and Mad2 were components of the same checkpoint pathway (51). Other researchers were able to provide insight into the function of BubR1 showing that once localized to the outer kinetochore plate, BubR1 binds to CENP-E and Bub3 (20, 52). This localization event is dependent on the previous localization of Bub1 and in the event Bub1 is missing or unable to bind to the kinetochore, the amount of BubR1 localized to the kinetochore is greatly reduced. Finally, it was observed that BubR1 must first interact with Bub3 before it could bind to the kinetochore (53, 54). BubR1 activation and mitotic spindle checkpoint: In the event of tension loss or microtubule instability during mitosis, BubR1 is hyper-phosphorylated and becomes active. This phosphorylation event is facilitated by Mad1 on the kinetochore and serves to alter BubR1 in such a way that it is able to interact directly with the APC as part of the mitotic checkpoint complex made up of the other checkpoint proteins Mad2 and Bub3 along with the APC cofactor Cdc20 or its subcomplexes (55). Without the presence of Mad1, while still on the kinetochore, and Mad2, BubR1 would not be able to interact with Cdc20 (53). This interaction inhibits the activity of the APC by preventing meaningful interactions between the APC and its substrates (20, 45, 52, 53). In all of these studies, it was shown that the BubR1 kinase domain was not necessary for the direct stoichiometric binding of BubR1 or the mitotic 21

33 checkpoint complex (MCC) to inhibit the APC. In addition to direct binding of the MCC to inhibit the APC, Yoon et al. observed that BubR1 was able to phosphorylate Cdc20 in vitro and that this phosphorylation event prevented the interaction between Cdc20 and the APC, effectively leading to a metaphase/anaphase block (45). Additional functions of BubR1: Another checkpoint protein phosphorylated by BubR1 is the breast cancer susceptibility gene BRCA2. The downstream effects of this phosphorylation event are still not clear but it is suspected that this allows interaction of BRCA2 with its substrates leading to cell cycle arrest (52). In a study by Shin et al., it was suggested that the role of BubR1 activation may not just lie in its activities during spindle checkpoint activation but also that the prolonged activation of the spindle checkpoint through BubR1 could promote apoptosis in those cells that eventually adapted to the arrest and exited mitosis, effectively executing a fail safe mechanism to prevent the propagation of cells breaching the mitotic checkpoint. This induction of apoptosis was found to be through the intrinsic pathway and the activation of caspase-9 leading to the subsequent activation of caspase-3 and apoptosis. A study by Vogel et al. further showed that cells need a functional mitotic checkpoint to activate post mitotic G1 arrest (56). Finally, Lens et al. found that for BubR1 to sustain its activity during prolonged checkpoint activation, the protein Survivin was necessary. It was also suggested that BubR1 was a possible substrate for Aurora B Kinase (47, 57). In addition to its role in the Spindle checkpoint, BubR1 has been found to interact with Beta2-adaptin, a subunit of AP2, on its N-terminal trunk domain and the C-terminal kinase domain of BubR1. AP2 is a member of the 22

34 assembly protein family involved in vesicular transport through the cell. AP2 mediates rapid endocytosis of the plasma membrane. It was found that BubR1 is able to interact with B2-adaptin through out the cell cycle suggesting that BubR1 might have a novel role in the regulation of vesicular intracellular traffic by regulating the soluble pools of B2- adaptin (46). Finally, in recent work by Fang et al. it was observed that MEF cell lines deficient for BubR1 exhibited compromised mitotic arrest and DNA repair after DNA damage by UV or the drug doxorubicin. DNA repair was found to be compromised in these cell lines through the down-regulation of p53, p21, phospho-h2ax and the enhanced degradation of PARP-1. Taken together this strongly suggests that a deficiency of BubR1 leads to continued cell cycling even after exposure to DNA damaging agents and through the down-regulation of key components of the DNA damage repair pathways, allows for the increased possibility of DNA damage and tumorigenesis (58) Effects of BubR1 inactivation: As it has been shown, BubR1 is an important component of the spindle checkpoint and has shown promise to play a role in several other checkpoints and cellular functions as well. BubR1 became a gene of interest due to the fact that mutant forms of the gene were observed in two colorectal cancer cell lines and that these cell lines exhibited chromosomal instability. From this it was suggested that the aneuploidy seen in cancer was due to the loss of chromosomal stability and that this loss was reflected through the mutation or loss of BubR1 (43). The idea that mutation in BubR1 caused chromosomal instability and its mutation was a factor in tumorigenesis was supported by 23

35 the work of Oshima et al who found that in four T-cell leukemia/lymphoma cell lines, BubR1 was found to have two missense mutations, one nonsense mutation, and an internal deletion (59). Furthermore, in breast cancer patients the aberrant expression of the protein encoded by the breast cancer specific gene 1 (BCSG1) has been shown to target BubR1 for degradation (60). With the ability to create knock-out mice using the gene trapping method, BubR1 haplo-insufficient mice were able to be bred to examine the effects of the loss of one BubR1 allele. By pairing mice both missing one functional allele of BubR1, it was determined that BubR1 was an essential gene and that embryos nullizygous for BubR1 begin to die 6.5 days after conception and that by 8.5 days, the nullizygous embryos are all reabsorbed. It was also determined that this embryonic death was due to extensive apoptosis (61, 62). The heterozygous mice lived well into adulthood but still exhibited problems and increased susceptibility to carcinogens. It was found that the spleens of heterozygous mice were enlarged with an increase in the number of mature megakaryocytes. These mice exhibit anemia and defects in platelet formation. It was hypothesized that this increase in spleenic megakaryocytes was due to the loss of proper cell division that in turn resulted in the formation of multinucleated megakaryocytes (62). It was also observed that mice heterozygous for the BubR1 gene trap were prone to tumors after exposure to azoxymethane, a known colon carcinogen. Heterozygous mice exposed to carcinogens develop colon masses two months after exposure while those with both functional alleles did not develop masses until six to eight months after treatment. It was also found that the heterozygous mice also developed tumors of the lung and liver, that was not observed in the wild-type mice (61). Later, Baker et al were 24

36 able to produce a hypomorphic mouse that expressed only 11% of the wild-type BubR1 level (BubR1 heterozygous mice express ~25% of wild-type levels). These mice were able to survive to adulthood but showed severe phenotypes including cachexia and lordokyphosis from three to six months of age and had a median lifespan of only six months, compared to the approximate two years observed in wild-type mice. These mice were also found to be infertile. Embryonic fibroblasts of BubR1 hypomorphic mice showed a severely compromised spindle checkpoint and a high level of aneuploidy. It was suggested in this study that BubR1 had a role in aging as those mice hypomorphic for BubR1 exhibited symptoms of advanced age (63). In work done by Rao et al. it was observed that BubR1 insufficiency had a synergistic effect with other proliferative control genes in the progression of tumorigenesis. It was observed that in a mouse strain exhibiting a mutation of the adenomatous polyposis coli gene (Apc), the introduction of BubR1 insufficiency led to a ten fold increase in the incidence of tumors in the colon. It was also seen that these tumors were in advanced stages of development. It was hypothesized that this increased tumor incidence was due to the role BubR1 played to prevent chromosomal instability. (64) Rational for Generating BubR1/p53 Compound Mutants: In this study, we have analyzed the synergistic effects of p53 deletion in the tumorigenesis in mice deficient for BubR1. Studies have shown that cancer is a multistep process that requires the inactivation of a variety of gene products involved in proliferation control. Furthermore, Fearon and Vogelstein and colleagues have shown the necessity for p53 inactivation in human colon cancer patients (65). Based on these 25

37 observations we hypothesized that the additional inactivation of the p53 tumor suppressor gene will either enhance the tumor susceptibility, indicated by a decrease in the lifespan, of the BubR1 mutant mice and/or induce the formation of colon cancer. To measure tumor susceptibility we generate a compound mutant for BubR1 and p53 and monitored the colony for a period of two and a half years. 26

38 II. Materials and Methods: Generation of BubR1 Mutant Mice in a p53 Deficient Background: In order to study the role of BubR1 in the spindle assembly checkpoint and cancer formation, BubR1 deficient mice were generated using embryonic stem (ES) cells obtained from the Mutant Mouse Resource Center at UC Davis. A gene trap approach was used to achieve disruption of the BubR1 gene as described earlier (66). The gene trap, containing a splice acceptor and an ATG-less betagalactosidase-neomycin fusion cassette, was determined to have inserted into intron 2 of the BubR1 gene. The genetrapped ES cells, from the 129P2/ OlaHsd strain, were analyzed by PCR and Southern blotting to confirm the presence of the trap within the BubR1 gene. Upon confirmation, the mutant ES cells were used for injection into 3.5 day old blastocysts and implanted into a pseudo-pregnant c57 bl/6 mouse. Nine chimeric mice were generated from these blastocyst injections and of these, four founder males produced germ line litters for the BubR1 deletion as detected by the agouti coat color. Tail clips were taken from the first two litters for genotype analysis. These analyses indicated equal numbers of wild type and heterozygous offspring consistent with the expected Mendelian ratio. Additionally, the heterozygous mice had no recognizable developmental defects. Heterozygous F1 offspring from the chimera males were then crossed in an effort to obtain BubR1 nullizygotes. To generate compound heterozygotes, the BubR1 mutant mice were crossed to our p53 deficient mouse strains (67). 27

39 DNA Extraction: To obtain DNA from our mice for Southern blotting and PCR based genotyping, we first took tail clippings from the pups during weaning and incubated each tail separately in 500 µl of tail lysis buffer (50 mm Tris ph 7.5, 50 mm EDTA, 100 mm NaCl, 1% SDS) along with 10 µl of Proteinase K and 1.25 µl 2M DTT. These tail clippings were then incubated overnight at degrees Celsius. After incubation, 500 µl of phenol: chloroform (24:1) was added and the contents were mixed well. The lysates were then spun in a microcentrifuge for three minutes at 13,000 rpm. Approximately 400 µl of the top layer (the aqueous phase) are removed into a second tube. After collecting the aqueous phase, twice the volume (800 µl) of 100% ethanol is added to each sample. Genomic DNA was precipitated out of solution and spooled out and placed into a third tube containing µl of 5 mm Tris (10mM Tris, 1mM EDTA) depending on the amount of DNA recovered. DNA was solubilized overnight and concentrations determined for use in southern blotting and PCR genotyping. BubR1 Genotyping: To determine the BubR1 genotype, we employed a PCR based strategy. For each sample, 200 ng of DNA was added to 23 µl of a PCR master mix (containing 1X PCR Buffer, 1.5 mm MgCl 2, 0.2 mm dntp (each), 0.2 µm of the forward and reverse primers, 1.0 unit of Platinum Taq DNA Polymerase). The PCR conditions for the insertion and wild-type PCR were as follows: an initial two minute denaturation step at 94 o C, followed by another denaturation step for 30 seconds at 94 o C, annealing for 30 28

40 seconds at 54 o C followed by an extension step for 135 seconds at 72 o C. The second denaturation, annealing, and extension steps were repeated for 32 cycles and then a final five minute extension step at 72 o C. For the Trap PCR the conditions were essentially the same with a decrease in extension time (75 seconds). The primers used to determine if the trap was located in the DNA were TR2 (5 CAACACTTGTATGGCCTTGGCG-3 ) and TR3 (5 GTGAGCGAGTAACAACCCGTC-3 ). This created a PCR product of 665 bp and verified that the trap was located in the sample. To determine the proper insertion of the gene trap, the primers GS1 (5 TTGGCAAAGCAAGAGTCAGC-3 ) and TR1 (5 CCCAACTGACCTTGGGCAAGAACATA-3 ) were used, creating a PCR product of 2.4 kb. These two products together verified the presence and proper insertion of the gene trap in our BubR1 gene. To determine if the wild-type allele was there we used the primers GS1 and GS2 (5 CCAGCCTAGGATACTTGGAGA-3 ). This produced a PCR product of 2.2 kb and indicated the presence of the undisturbed intron. Generation of Mouse Embryonic Fibroblasts (MEF): Compound heterozygous crosses were performed and the pregnant female sacrificed at embryonic day 13.5 and the embryos removed and placed into separate 15 ml test tubes. Using a two ml syringe with a small bore needle, the embryo was placed into the syringe and pressed through the needle into cell media (DMEM, 15% FBS, 10 units of Pen/Strep). The cell suspension was then added to 2 large plates and incubated for two to three days. Once confluent, the plates were split 1:3 and allowed to grow until confluent and then split again at a ratio 1:3. A portion of the cells were isolated and used for experiments, while the rest of the cells were frozen in liquid nitrogen. 29

41 Cell Proliferation Assay: Cell lines at the same passage were removed from liquid nitrogen storage and plated on 60 mm culture plates in media (DMEM, 15% FBS, 10 units of PenStrep). At 90% confluence, cell were trypsinized, and re-plated, onto a 100 mm plate, and allowed to grow to 80% confluence. Once confluent, these plates are then trypsinized and counted (to count the cells, 50 ul of the collected cell suspension was added to 100 ul of PBS/EDTA and from that mixture 50 ul of the solution was placed onto a hemocytometer and the four blocks counted, divided by four and multiplied first by the dilution factor and then by 10,000 to get the number of cells per milliliter of cell suspension) and 125,000 cells are seeded onto two 6-well plates and allowed to grow for one day. On day two and each of the following five days, two wells from each cell line are trypsinized and counted. These numbers are then plotted. Colony Forming Unit Assay (CFU): Cell lines were removed from liquid nitrogen and seeded onto 60 mm plates with media (DMEM, 15% FBS, 10 units of PenStrep) and then passage on a three-day/fourday cycle until passage six, seeding 400,000 cells each time. At passage six, the plates were trypsinized and counted and 400,000 cells were passaged on to passage 12 where the second part of this study was done. At passages six and twelve, 2,500 cells were seeded onto each of four small plates and allowed to grow for nine days with media changes every three days. After this growth period, the media was removed from each plate, rinsed with PBS, and 100% 30

42 methanol was added to each plate to fix the cells for nine minutes. After fixing the cells, Giemsa stain (1.25 ml Giemsa stain, 1.5 ml 100% Methanol, QS to 50 ml) was added to cover the cells and allowed to incubate for 9 minutes. After this incubation, the stain was removed and each plate was rinsed twice with distilled water to remove any access stain remaining in the plate. After drying, colonies consisting of 40 cells or more were considered in the colony count. Cell Cycle Analysis: Low passage MEFs cell lines were seeded on to 35 mm tissue culture plates (300,000 cells/plate) and were allowed to grow for ~24 hours before the addition of the spindle poison, nocodazole (125ng/ml). Cells were trypsinized at various time points (8, 24 and 48 hours) and collected by centrifugation. Single cell suspensions were prepared from the cell pellets using a vortex, fixed in 70% ethanol, and stored at -20 o C until the DNA content could be examined via flow cytometry for each genotype. Animal Husbandry: Mice were housed in small sterilizable cages with isolation lids and fed ad lib. At 21 days, pups were weaned and a tail clipping was taken for DNA extraction. In this controlled environment, mice were examined 3 times weekly for signs of illness or tumor formation. In the event a mouse was found to exhibit morbidity or have developed a palpable mass, it was euthanized using carbon dioxide and dissected. During autopsy, tumor tissues and organs exhibiting abnormal morphology were removed and a sample of 31

43 it was formalin-fixed and another sample was removed and stored at -80 o C to be used for genotyping. Southern Blotting: To determine the p53 genotype of the mice, southern blot assays were performed. Genomic DNA (3 µg) from the tail clipping was digested with Bam HI overnight at 37 o C. The next day the samples were subjected to electrophoresis, buffered in tris-acetate- EDTA (40mM tris base, 20 mm acetate, 2mM EDTA) and imaged. After imaging, the membrane was washed in 0.25 M HCl for 5 minutes, rinsed 4 times with distilled water and transferred onto a nylon membrane overnight using 0.4 N NaOH. Following DNA transfer, the membrane containing the blotted DNA was rinsed with 2X SSC, dried at 37 o C and blocked with pre-hybridization buffer (1.3% SDS, 2X SSPE, 1% milk, and 2 mg denatured salmon sperm DNA) for 5 hours at 68 o C. After blocking, the prehybridization buffer was removed and 20 ml of hybridization buffer was added (10% dextran sulfate, 1.5X SSPE, 1% SDS, 0.5% milk) along with the radio-labeled probe. (1x10 6 cpm/ml). After incubating overnight, the hybridization buffer was removed and the membrane was rinsed with 2X SSC followed by a series of solutions (Solution 1: 2X SSC, 0.1% SDS. Solution 2: 0.5X SSC, 0.1% SDS. Solution 3: 0.2X SSC, 0.2% SDS). Once the membrane activity is between 2,000 and 3,000 counts per minute, the membrane was exposed on x-ray film overnight at -80 o C. This was to reduce the amount of nonspecific binding of the radioprobe to the membrane, allowing for clearer results. The radio-labeled probe was generated using 100ng of DNA specific to exons 2-6 (a 600 bp fragment from the Kpn I digested plasmid (LR10) containing murine p53 cdna) of 32

44 the p53 gene. This probe was then labeled with dctp[α- 32 P] using random primers. The specific activity of the radiolabel was 3,000 Ci/mmol. In addition to the two possible bands that represent the wild type and mutant alleles, the p53 probe also binds to pseudogene regardless of the p53 genotype (figure 4). 33

45 Pseudo-gene Mutant Allele Wild-Type Allele Figure 4: Southern blot results indicate the p53 genotype. An image of the southern blot with the bands labeled for 6 samples (from genomic DNA extracted from the tails of mice). The top band is a pseudogene, the middle band indicating the mutant allele (in which a small portion of exon 5 was deleted and a neomycin fusion cassette inserted) and the bottom band indicates the presence of the wild type allele. 34

46 III. Results: Characterization of the BubR1 gene Trap Insertion Site: The 5 RACE data obtained from the Mutant Mouse Regional Resource Center (MMRRC) indicated that the trap had inserted into intron 2 of the gene. Using this data, we designed primers specific for exon 2 of BubR1 (GS1) and the gene trap (TR1). PCR amplification produced a 2.4 Kb fragment that was specific only to the trapped allele (as determined by the trap-specific PCR primers TR2 and TR3). Sequence analysis of the PCR product revealed that the trap had inserted into intron 2 of the BubR1 gene at base 2,170. Integration of the trap also resulted in the loss of the first 880 base pairs within the trap. However, this did not affect the function of the trap since the splice acceptor lies outside these lost bases (~1,276). Based on this information, we designed primers adjacent to the insertion site to amplify a 2.2 Kb fragment specific for the wild-type allele. A schematic representation of the primer sites with respect to the gene trap are shown in figure 5. An example of the PCR products obtained for various combinations of primer pairs is shown in figure 6. BubR1 Deficiency Affects Embryogenesis: We generated F1 heterozygotes using male chimeras and wild type females. The BubR1 heterozygotes developed normally and were capable of producing progeny. In order to generate mice that were completely deficient for the BubR1 allele we 35

47 Wt allele GS1 2.1kb GS2 3 kb Trapped allele TR1 TR2 TR3 1 2 β-gal-neo Trap (7.7kb) 3 4 Figure 5: The BubR1 gene and the location of the PCR primers. This schematic representation of the wild-type and trapped alleles show the positions of the PCR primers used for genotyping. In addition to this the blue triangle indicates the location where the gene trap inserts into the BubR1 gene. 36

48 12,000 Marker Blank well , Size (bp) Trap PCR (665 bp) Insertion PCR (2.4 kb) Wild-Type PCR (2.2 kb) Figure 6: PCR indicates the BubR1 genotype. This gel image provides BubR1 genotype data for 5 samples. The marker here is a 1 kb+ ladder. The trap PCR used the primers TR2 and TR3 from the schematic above while the Insertion and wild type PCRs used the primer GS1 and either TR1 (the insertion) of GS2 (the wild type). The trap and insertion PCR indicate the presence of the trapped allele (samples 1, 2 and 4). Absence of a band indicates a wild type genotype (see samples 3 and 5). The wild-type PCR indicates the presence of the wild-type allele. Absence of a band would signify the loss of the wild-type allele. 37